Mathematical Reasoning on AIME25 (Accuracy MV, Accuracy WMV, Required Tokens)
83.3Accuracy (MV)ASC
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| ASCBackbone=DeepSeek-8B, branching factor n=322026.06 | 83.3 | — | 12 | |
| DeepConfBackbone=DeepSeek-8B, branching factor n=322026.06 | 83.3 | 83.6 | 18.9 | |
| Flex-Fork-thinkBackbone=DeepSeek-8B, branching factor n=32, seed path length l0=2,0482026.06 | 82.4 | 83.3 | 11.6 | |
| DeepConfBackbone=Phi-4-RP-14B, branching factor n=322026.06 | 76.3 | 76.3 | 11.2 | |
| Flex-Fork-thinkBackbone=Qwen3-8B, branching factor n=32, seed path length l0=2,0482026.06 | 75.6 | 74.4 | 9.1 | |
| ASCBackbone=Qwen3-8B, branching factor n=322026.06 | 75.4 | — | 11.6 | |
| DeepConfBackbone=Qwen3-8B, branching factor n=322026.06 | 69 | 71 | 13.8 | |
| Flex-Fork-thinkBackbone=Phi-4-RP-14B, branching factor n=32, seed path length l0=2,0482026.06 | 64.6 | 67.3 | 8.3 | |
| ASCBackbone=Phi-4-RP-14B, branching factor n=322026.06 | 62.6 | — | 13.5 |